pubs.acs.org/jmc Published on Web 01/31/2011 r 2011 American Chemical Society

1080 J. Med. Chem. 2011, 54, 1080–1090

DOI: 10.1021/jm101566d

Biotin Ergopeptide Probes for Dopamine Receptors

Marc Vendrell,†,^ Anabel Molero,†,§ Sergio Gonz�alez,‡ Kamil P�erez-Capote,‡ Carme Lluis,‡ Peter J. McCormick,‡

Rafael Franco,‡ Antoni Cort�es,‡ Vicent Casad�o,‡ Fernando Albericio, ),§ and Miriam Royo*,†

†Combinatorial ChemistryUnit, Barcelona SciencePark,University of Barcelona, 08028Barcelona, Spain, ‡Centro de Investigaci�onBiom�edica enRed sobre Enfermedades Neurodegenerativas (CIBERNED), and Department of Biochemistry and Molecular Biology, Faculty of Biology,University of Barcelona, 08028 Barcelona, Spain, §CIBER-BBN, Networking Centre on Bioengineering, Biomaterials, and Nanomedicine,Barcelona Science Park, 08028 Barcelona, Spain, and )Institute for Research in Biomedicine, 08028 Barcelona, Spain. ^Current address:Laboratory of Bioimaging Probe Development, Singapore Bioimaging Consortium, 138667 Singapore

Received December 8, 2010

The incorporation of chemical modifications into the structure of bioactive compounds is often difficultbecause the biological properties of the newmoleculesmust be retainedwith respect to the native ligand.Ergopeptides, with their high affinities at D1 and D2 dopamine receptors, are particularly complexexamples. Here, we report the systematic derivatization of two ergopeptides with different peptide-based spacers and their evaluation by radioligand binding assays. Selected spacer-containing ergopep-tides with minimal biological alteration and a proper anchoring point were further derivatized with abiotin reporter. Detailed characterization studies identified 13 as a biotin ergopeptide maintaining highaffinity and agonist behavior at dopamine receptors, being a useful tool for the study of heteromersinvolving D1R, D2R, or D3R.

Introduction

Ergopeptides, with their high affinity at D1 and D2 dopa-mine receptors (D1Rs and D2Rsa), are valuable molecules tostudy dopamine receptors.1 The therapeutic significance ofdopamine receptors containing heteromers has been extensively

reported.2-4 Dual ligands have been successfully appliedto the study of G-protein coupled receptors (GPCRs)oligomerization.5,6 While ergopeptides may provide insightsinto the D1 orD2 receptors containing heteromers, they requirea proper reporter tag prior to their use in protein localizationand profiling studies.7-10 The modification of hit compoundsis a critical step as the biological properties of the labeledmolecules canbe significantly alteredwith respect to thenativeligand.11 This step is particularly important when applied tosmall molecule ligands, such as ergopeptides, because theligand affinity and efficacy of the pharmacophore is likely tobe compromised.12 In thepresentwork,we report a systematicstudy to optimize the length and chemical nature of differentspacer moieties attached to two ergopeptides with high affi-nity at D1R and D2R (1 and 2, Chart 1) and identify thoselinkers that retain their binding profile. Trifunctional aminoacids, such as lysine and glutamic acid, are excellent scaffoldsfor the synthesis of peptide-based spacers. In addition to theirlow toxicity, they can be easily adapted to a solid-phasesynthesis approach to allow the incorporation of a range offunctional groups (amines, anilides, and carboxamides) with-in the spacer structure. A number of peptide-based spacerswere incorporated to 1 and 2 to render a 40-member library ofnew ergopeptides, and the evaluation of their affinities atD1Rand D2R identified two linker moieties with minimal biologi-cal interference. The subsequent incorporation of a biotinreporter led to identification of 13 as a biotin ergopeptide fordopamine receptors with nanomolar binding affinities andagonist behavior.

Results

Design of the Library. The incorporation of a spacermoiety into a bioactive molecule requires the selection of asuitable attachment point, which must be well separated

*To whom correspondence should be addressed: Phone: 0034934037122. Fax: 0034 934037126. E-mail: mroyo@pcb.ub.cat.

aAbbreviations: A2AR, adenosine A2A receptor; A1R, adenosine A1

receptor; Ac2O, acetic anhydride; ACN, acetonitrile; Alloc, allyloxy-carbonyl; All, allyl ester; Ahx, aminohexanoic acid; Akt, protein kinaseB; Boc, t-butoxycarbonyl; CDI, 1,10-carbonyldiimidazole; CHO,Chinese hamster ovary; DCM, dichloromethane; cDNA, cDNA; [3H]-CP55940, tritium labeled 2-[(1R,2R,5R)-5-hydroxy-2-(3-hydroxypropyl)-cyclohexyl]-5-(2-methyloctan-2-yl)phenol; DIPCDI, N,N0-diisopropyl-carbodiimide; DIEA, diisopropylethylamine; DMF, dimethylforma-mide; DPCPX, 8-cyclopentyl-1,3-dipropylxanthine; D1R, dopamineD1 receptor; D2R, dopamine D2 receptor; D3R, dopamine D3 receptor;ERK, extracellular-signal-regulated kinases; Fmoc, fluorenylmethylox-ycarbonyl; GPCR, G-protein coupled receptor; HEPES, 4-(2-hydro-xyethyl)-1-piperazineethanesulfonic acid;HOBt, hydroxybenzotriazole;HOAt, 1-hydroxy-7-azabenzotriazole; HRMS, high resolution massspectrometry; MAPK, mitogen-activated protein kinases; PEG, poly-ethyleneglycol; PEI, polyethylenimine;Rink-MBHA-PS, 4-(20,40dimethoxy-phenyl-Fmoc-aminomethyl)-pheoxyacetamido p-methylbenhidryla-mine resin; RP-HPLC, reversed phase-high performance liquid chro-matography; RP-HPLC-MS, reversed phase-high performance liquidchromatography-mass spectroscopy; SKF81297, (()-6-chloro-2,3,4,5-tetrahydro-1-phenyl-1H-3-benzazepine hydrobromide; TFA, trifluor-oacetic acid; TMUCl Cl, N-[chloro(dimethylamino)methylene]-N-methylmethanaminium chloride; TBTU, O-(benzotriazol-1-yl)-N,N,N0,N0-tetramethyluronium tetrafluoroborate; Tris-HCl, tris(hydroxy-methyl)aminomethane hydrochloric acid; [3H]-RAMH, tritium labeledR-methyl histamine; [3H]-R-PIA, tritium labeled R-phenylisopropyla-denosine; [3H]-SCH23390, tritium labeled [R-(þ)-8-chloro-2,3,4,5-tet-rahydro-3-methyl-5-phenyl-1H-3-benzazepine-7-ol); [3H]-YM09151-2,tritium labeled nemonapride (N-(1-benzyl-2-methylpyrrolidin-3-yl)-5-chloro-2-methoxy-4-(methylamino)benzamide); [3H]-ZM241358, tri-tium labeled 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl)phenol. Abbreviations used for amino acids follow theIUPAC-IUB Commission of Biochemical Nomenclature in Jones, J. H.J. Pept. Sci. 2003, 9, 1-8.

Article Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 1081

from the pharmacophore in order to prevent any interferenceon the binding between biomolecule and receptor. Thebinding mode analysis of 1 and 2 revealed the interactionof the ergolene scaffold at the transmembrane binding siteswhile the peptide moieties interacted with adjacent aminoacids.1 Bearing this inmind, the attachment of distinct spacermoieties into 1 and 2 was designed by means of the C-term-inal carboxamide group modification, which would presum-ably produce a minor change to the biological properties ofthe final compounds.

Several spacer moieties were designed on the basis of twotrifunctional amino acids, lysine and glutamic acid, whichare key scaffolds for the derivatization of biomolecules whena suitable protecting group scheme is used.13 The introduc-tion of a number of building blocks onto the lysine andglutamic acid side chains afforded 19 spacer moieties withvarious length and chemical functionalities (Table 1), includ-ing aromatic and saturated rings, primary amines, carbox-amides, anilides, and poliethylenglycol (PEG) units. Moreover,a common NR-group at the C-terminus was acetylated tomimic the further incorporation of a reporter molecule.

Synthesis of the Spacer-Containing Ergopeptides. The synthe-sis of thepeptide-based spacerswas entirely performedon solid-phase, using Rink-MBHA-PS as a polymeric support.Coupling Fmoc-Lys(Alloc)-OH or Fmoc-Glu(OAll)-OH ontothe resin, Fmoc elimination, acetylation of the NR-group, andremoval of the side chain protecting groups led to resins 3 and 4,which were used for the construction of the 19 peptide-basedspacers (Scheme S1 in Supporting Information (SI)).

The introduction of distinct units (amino acids, diamines,anhydrides, and dicarboxylic acids) onto resins 3 and 4 wascarried out using standard solid-phase peptide synthesis(SPPS) protocols and 1,10-carbonyldiimidazole (CDI) forthe activation of supported carboxylic acids. The poor nucleo-philicity of some anilines required the use of a previouslydescribed procedure, based on N-[chloro(dimethylamino)-methylene]-N-methylmethanaminium chloride (TMUCl Cl)as the activating reagent, for the solid-phase synthesis ofanilides14 (compounds 1p, 2p, 1r, and 2r). The incorporationof the 19 spacer moieties into the ergopeptides structure wascarried out in two steps: (1)NR-Boc-tripeptides 5 and 6 wereattached to the supported spacers in a convergent protocolinstead of a stepwise procedure to favor the parallelizationsynthetic process; (2) the crude mixtures released aftercleavage (TFA-H2O, 95:5) were further coupled to the D-lysergic acid in solution using DIPCDI and HOAt (Scheme 1).Although this method involved an extra step compared to

the direct coupling of ergopeptides onto the spacer moieties,the amount of the costly D-lysergic acid could be minimized.The purification of the whole library by semipreparative RP-HPLC afforded the 38 final products with excellent purities(1a-s and 2a-s, Table S1 in SI).

Biological Assays: Binding Properties of Spacer-Contain-

ing Ergopeptides. The binding properties of 1a-s and 2a-s

were assayed by displacement experiments of D1R or D2Rradiolabeled ligands (concentration indicated in the legendsof Figures 1 and 2) by 25 μM 1a-s or 2a-s. The bindingscreening at D1R and D2R (Figures 1 and 2) indicated asignificant decrease of binding affinity when aromatic ringswere includedwithin the spacermoieties (1p (mainly atD2R),2i, 2j, 2p, and 2r). In contrast, the incorporation of linearaliphatic spacers resulted in compounds showing a morefavorable binding, with slightly stronger interactions in thepresence of medium-length spacers (1c, 1f, 1n at D1R; 1l, 1n,2f, and 2k atD2R) and some shorter ones (1a, 1m atD1R; 1a,1m, 2a at D2R). Regarding the inclusion of primary amineswithin the spacer structure, the results were both ergopeptideand receptor-dependent; 1d did not show a significantly differ-ent binding at D1R, but 1d and 1e binding affinities wereremarkably improved at D2R. Whereas determining theexact nature of this enhancement would require further studies,the incorporation of medium-sized aliphatic spacers (c and f,Table 1) at the C-terminus of ergopeptides 1 and 2 provedto be successful in preserving their binding affinities at bothD1R and D2R.

We further investigated the behavior of the modifiedergopeptides (1c, 1f, 2c, and 2f) at A1 and A2A adenosinereceptors because heteromers containing adenosine-dopa-mine receptors have been very well described.15-17 The bindingaffinities of 1c, 1f, 2c, and 2f were assayed by displacementexperiments of A1R or A2AR radiolabeled ligands (con-centration indicated in the Figure S1 and S2 legends in SI)by 25 μM ergopeptides. The binding of 1c, 1f, 2c, and 2f atadenosine receptors (Figures S1 and S2 in SI) proved to bemuch lower than at D1R and D2R and confirmed theirspecificity for dopamine receptors.

Synthesis and Pharmacological Characterization of Biotin

Ergopeptides (9-16). Biotin has been extensively used forprotein profiling studies because its tight interaction withavidin can facilitate the isolation of protein complexes.18-20

Moreover, the availability of numerous biotin/streptavidin-labeled antibodies and reagentsmakes biotin one of themostversatile reporters for protein characterization. To constructthe corresponding biotin ergopeptide probes, a biotin molecule

Chart 1. Ergopeptides with High Affinity at D1 and D2 Dopamine Receptors

1082 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 Vendrell et al.

was incorporated to the C-terminus of 1c, 2c, 1f, and 2f.Biotin ergopeptides were synthesized using a slightly mod-ified procedure from the previously described (Scheme 2),and the biotin reporter was coupled to the NR-group of theC-terminus of ergopeptides using TBTU and HOBt.

The larger size of the biotin ergopeptides facilitated theisolation of the two ergolene diastereomers derived from theepimerization of D-lysergic acid under the coupling con-ditions,21 and eight biotin ergopeptides (9-16, Chart 2) weresubjected to primary radioligand binding assays at D1R and

Table 1. Chemical Structures of Ergopeptides after the Incorporation of Peptide-Based Spacers

Article Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 1083

D2R.The binding properties of 9-16were initially evaluatedusing displacement experiments of a fixed concentration ofD1R, D2R, A1R or A2AR radiolabeled ligands (indicatedin the legends of Figures S3-S6 in SI) by 25 μM 9-16

(Figures S3-S6 in SI). The incorporation of the biotinmolecule retained the binding properties of the parent com-pounds 1 and 2 and confirmed that both selected spacers(c and f) separate well the pharmacophore from the reportertag. We selected compounds 9, 12, 13, and 14 for theirpharmacological characterization, and their affinity con-stants at D1R and D2R were calculated by competitionexperiments: brain membranes (0.5 mg/mL) were incubatedwith a fixed concentration of D1R or D2R radiolabeledantagonists in the absence or in the presence of increasingconcentrations of compounds 9, 12, 13, or 14, as described inthe Experimental Section. From the resulting competitioncurves (Figure 3 for compound 13, and Figures S7-S9 in SIfor compounds 9, 12, and 14), their correspondingKD valueswere determined (Table 2). The biotin ergopeptide 13 provedto be the compound that best maintained the nanomolarrange affinities at both dopamine receptor subtypes.

To test whether the compound 13 behaved as an agonist,we examined two different signal transduction pathways(e.g., MAPK and Akt (PKB)) in cells that were separatelytransfected with D1R and D2R. As shown in Figure 4a,b, 13increased the ERK1/2 phosphorylation in cells expressing

D1R or D2R in a dose-dependent manner and to a similarextent than a D1R full agonist (e.g., SKF 81297)22 or a D2Ragonist (e.g., quinpirole).23 The 13-mediated effect (at 0.1 or1 μM) was also reverted when cells were preincubated with 5or 50 μMofD1R (e.g., SCH 23390) or D2R antagonists (e.g.,raclopride), further demonstrating the agonist behavior of13 at both dopamine receptor subtypes. Regarding the Akt(PKB) pathway, both SKF 81297 and quinpirole (D1R andD2R agonists, respectively) induced a decrease in the AktSer473 phosphorylation, a signaling that has been also observedin mouse brain.24,25 Similarly, 13 decreased the Akt phosphor-ylation in a dose-dependent manner in cells expressing D1Ror D2R, and its effect was also reverted upon preincubationwith D1R (e.g., SCH 23390) or D2R antagonists (e.g.,raclopride), confirming the agonist behavior of 13 at bothD1R and D2R (Figure 4c,d).

Since it has been described that D1Rs can also heteromer-ize with dopamine D3Rs,26 we analyzed the binding affinityof compound 13 at D3R. Because of the low expression levelof D3R in the striatum when compared to D2R, we transi-ently transfected CHO cells with D3R to study the bindingaffinity of compound 13 to this receptor. Membranes(0.5 mg/mL) from transfected cells were incubated with4 nM [3H]-raclopride, a D3R antagonist, in the absence orin the presence of increasing concentrations of compound 13.The competition curvewasused todetermine the corresponding

Scheme 1. Synthesis of Representative Compounds 1c and 2s

aConditions: (a) Fmoc-Ahx-OH, DIPCDI/HOBt; (b) piperidine-DMF (2:8); (c) 5, DIPCDI/HOBt; (d) TFA-H2O (95:5); (e) D-lysergic acid,

DIPCDI/HOAt, DIEA; (f) (Fmoc-4-amino) piperidine 3HCl, DIEA, CDI; (g) piperidine-DMF (2:8); (h) 6, DIPCDI/HOBt.

1084 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 Vendrell et al.

KD values (KDB1 = 23 ( 9 nM and KDB2 = 93 ( 36 nM),proving that 13 can be a useful tool for the study of hetero-mers involving D1, D2, or D3 dopamine receptors (Figure 5).The binding properties of 13 were also examined at otherGPCRs that can form heteromers with D1Rs or D2Rs (e.g.,histamine H3, metabotropic glutamate 5, somatostatin SST5,and cannabinoids CB1 receptors).27-31 Displacement experi-ments (Figure S10 in SI) showed that high concentrations of13 did not significantly decrease the radioligand binding atthe studied receptors and proved that the binding of 13 atother GPCRs was very low when compared to dopaminereceptors.

Conclusions

The derivatization of two ergopeptides showing high affi-nity at dopamine receptors has been optimized using a com-binatorial chemistry approach to develop of a novel biotinergopeptide that maintained both nanomolar binding affi-nities and an agonist behavior at dopamine receptors. Thesystematic modification of the two parent ergopeptides usinga solid-phase synthesis approach afforded a 40-member li-brary including different peptide-based spacers at the C-ter-minus of the ergopeptides. The binding analysis of the libraryidentified two modified ergopeptides incorporating medium-length aliphatic spacers as the compounds that best retainedthe affinity profile at D1R and D2R. Subsequent derivatiza-tion of the spacer-containing ergopeptides with a biotin mole-cule rendered a set of biotin ergopeptides that bound at D1Rand D2R with KD values in the nanomolar range. Furthercharacterization studies identified 13 as a biotin ergopeptide

with agonist behavior and significantly higher binding affi-nities at D1R, D2R, and D3R when compared to adenosine(A1/A2A), histamine H3, metabotropic glutamate 1/5, soma-tostatin SST, and cannabinoid CB1 receptors. These resultsattest the potential application of 13 to study heteromer com-plexes involving dopamine receptors. Experiments with biotinergopeptides to study dopamine receptors heteromers arecurrently ongoing and will be reported in due course.

Experimental Section

Materials and Equipment. All Fmoc-amino acids were pur-chased fromNeosystem (Strasbourg, France), and Fmoc-Rink-PS and 2-chlorotrityl resins were supplied by Calbiochem-NovabiochemAG.DIPCDIwas obtained fromFlukaChemika(Buchs, Switzerland) and HOBt from Albatross Chem, Inc.(Montreal, Canada). Solvents for peptide synthesis and RP-HPLC equipment were obtained from Scharlau (Barcelona,Spain). Trifluoroacetic acid was supplied by KaliChemie (BadWimpfen, Germany). Other chemicals of the highest commer-cially available purity were purchased from Aldrich (Milwaukee,WI). All commercial reagents and solvents were used as received.

Adenosine deaminase (EC 3.5.4.4) was purchased fromRoche (Basel, Switzerland), and [3H]-R-PIA was supplied byAmersham Biosciences (Buckinghamshire, UK). Raclopride,polyethylenimine (PEI), MgCl2, DPCPX, mouse antiphospho-ERK1/2 antibody, rabbit anti-ERK1/2 antibody, IRDye 800antimouse antibody, and IRDye 680 antirabbit antibodies werepurchased fromSigma (St Louis,MO). Rabbit anti-P-Ser473Aktantibody was purchased from SAB Signalway (Pearland, USA).ZM241385, SCH23390, RAMH, quisqualic acid, somatostatin,and CP55940 were supplied by Tocris Biosciences (Avonmouth,UK). [3H]-SCH23390, [3H]-YM09151-2, [3H]-quisqualic acid,

Figure 1. Displacement experiments atD1R. Specific binding of 0.9nM D1R antagonist [3H]-SCH23390 in the absence or in thepresence of competing ligands was measured as indicated in theExperimental Section. C, control of radioligand binding withoutcompeting ligand; 1a-2s, radioligand specific binding in the pre-sence of 25 μM ergopeptides; 1-2, radioligand specific binding inthe presence of 25 μMergopeptides 1 and 2, respectively. Values arerepresented as means ( SD (n = 3). Student’s t-test for unpairedsamples showed significant differences (**p < 0.01) compared tothe ergopeptides 1 or 2.

Figure 2. Displacement experiments at D2R. Specific binding of0.7 nM D2R antagonist [3H]-YM 09151-2 in the absence or in thepresence of competing ligands was measured as indicated in theExperimental Section. C, control of radioligand binding withoutcompeting ligand; 1a-2s, radioligand specific binding in the pre-sence of 25 μM ergopeptides; 1-2, radioligand specific binding inthe presence of 25 μMergopeptides 1 and 2, respectively. Values arerepresented means ( SD (n = 3). Student’s t-test for unpairedsamples showed significant differences (*p < 0.05; **p < 0.01)compared to the ergopeptides 1 or 2.

Article Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 1085

[3H]-CP55940, [125I]-Tyr11-somatostatin 14, and [3H]-ZM241385were supplied by Perkin-Elmer (Boston,MA). [3H]-RAMHwaspurchased fromGEHealthcare (Buckinghamshire, U.K). Ecos-cint H scintillation cocktail was purchased fromNational Diagnos-tics (Atlanta,GA).Bradfordassaykitwaspurchased fromBio-Rad(Munich, Germany). All other supplements were purchased fromInvitrogen (Paisley, UK).

Analytical RP-HPLC-MS was performed using 2795 Waters(Milford, MA) Alliance with a Micromass ZQ mass spectro-meter and a 996 PDA detector. Semipreparative RP-HPLCwasperformed on a 2767 Waters chromatography system with aMicromass ZQ mass spectrometer. Multiple sample evapora-tion was carried out in a Discovery SpeedVac ThermoSavant(Waltham, MA). Radioligand binding experiments were per-formed using a Brandel (Gaithersburg,MD) cell harvester and aPackard 1600 TRI-CARB scintillation counter. Fitting databinding program GRAFIT was obtained from Erithacus Soft-ware (Surrey, UK). For ERK1/2 or P-Ser473Akt phosphoryla-tion determination, the Odyssey infrared scanner (LI-CORBiosciences, Lincoln, Nebraska, USA) was used. Band densitieswere quantified using the scanner software and exported toExcel (Microsoft, Redmond, WA, USA).

Synthesis. Solid-Phase General Procedure. Peptide syntheseswere performed manually in a polypropylene syringes, eachfitted with a polyethylene porous disk. Solvents and solublereagents were removed by filtration. Washings between depro-tection, coupling, and subsequent deprotection steps were carried

out withDMF (5� 1min) andDCM (5� 1min) using 10mL ofsolvent/g of resin each time.

Coupling using DIPCDI and HOBt-HOAt. Fmoc-AA-OH,carboxylic acids or Boc-peptides (3 equiv) were coupled usingDIPCDI (3 equiv) as coupling reagent and HOBt (3 equiv) orHOAt (3 equiv) as additives in DCM-DMF (1:1) for 2-4 h at rt.After each coupling, the resinwaswashedwithDMF (5� 1min)and DCM (5 � 1 min). Reaction completion was checked bymeans of the Kaiser or chloranil tests.32

Anhydrides Coupling. Anhydrides (10 equiv) were coupledusing DIEA (10 equiv) in DCM for 2-4 h at rt. After eachcoupling, the resinwaswashedwithDMF (5� 1min) andDCM(5 � 1 min). Reaction completion was checked by means of theKaiser or chloranil tests.

Coupling using CDI.33 Solid-supported carboxylic acids were

washedwithDCM(5� 1min) andDMF(5� 1min) and treatedwith CDI (25 equiv) in DMF (30 min). After filtering the resinand washing with DMF, amines (5 equiv) were added in DCM-DMF (1:1) andkept under orbital agitation for 2-4 h at rt.Aftereach coupling, the resin was washed with DMF (5� 1 min) andDCM (5 � 1 min). Reaction completion was checked by meansof the malachite green test.34

Coupling using TMUCl Cl.14 Solid-supported carboxylicacids were washed with DCM (5 � 1 min) and treated withTMUCl Cl (10 equiv) and DIEA (10 equiv) in DCM (10 min).After filtering the resin and washing with DCM and DMF,anilines (5 equiv) were added in DCM-DMF (1:1) and left

Scheme 2. Synthesis of Representative Biotin Ergopeptides 11 and 13. a

aConditions: (a) Pd(PPh3)4-PhSiH3 in DCManh; (b) biotin, TBTU/DIEA, HOBt; (c) piperidine-DMF (2:8); (d) sequential peptide synthesis:

couplings in DIPCDI/HOBt, Fmoc removal in piperidine-DMF (2:8); (e) TFA-H2O (95:5); (f) D-lysergic acid, DIPCDI/HOAt.

1086 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 Vendrell et al.

under orbital agitation for 2 h at rt. After each coupling, theresin was washed with DMF (5 � 1 min) and DCM (5 �1 min).

Fmoc Group Removal involved the following sequence: (i)DMF (5 � 1 min); (ii) piperidine-DMF (2:8) (1 � 1 min þ 2 �15 min); (iii) DMF (5 � 1 min).

Alloc/All Group Removal involved the following sequence: (i)DCM (5 � 1 min); (ii) Pd(PPh3)4 (0.1 equiv) and PhSiH3

(10 equiv) in anhydrous DCM (3 � 15 min); (iii) anhydrousDCM (5� 1 min); (iv) DCM (5� 1 min); (v) DMF (5� 1 min);(vi) 0.02 M solution of sodium diethyldithiocarbamate in DMF(3 � 15 min), DMF (5 � 1 min), DCM (5 � 1 min), and DMF(5 � 1 min).

Nr-Terminus Acetylation. The resins were treated with Ac2O(10 equiv) andDIEA (10 equiv) in DCM (2� 15 min). Reactioncompletion was checked by the Kaiser test.

Cleavage Conditions. A. 2-Chlorotrityl-Based Resins.The resinswere treated with a solution of TFA-DCM (5:95) (5 � 1 min).Filtrates were collected, washed with DCM (3 � 1 min), andevaporated under vacuum.

B. Rink-Based Resins. The resins were treated with a solutionof TFA-H2O (95:5) and orbitally shaken for 2 h at rt. Filtrateswere collected, washed with TFA (2 � 1 min) and DCM (3 �1 min), and evaporated under vacuum.

Synthesis of Boc-Val-pNO2Phe-Pro-OH (5). Starting from2-chlorotrityl resin (6.0 g, loading: 1.3 mmol/g), the peptide wassynthesized as described above. Cleavage and lyophilizationin H2O-ACN (2:1) rendered a white solid powder; 1.75 g(yield: 39%); Mexp 507.1 (Mcalc 506.3), 96%.

Synthesis of Boc-pNO2Phe-pFPhe-Nip-OH (6). Starting from2-chlorotrityl resin (6.0 g, loading: 1.3 mmol/g), the peptide wassynthesized as described above. Cleavage and lyophilizationin H2O-ACN (2:1) yielded a white solid powder; 3.58 g (yield:68%); Mexp 587.0 (Mcalc 586.2), 95%.

Coupling of D-Lysergic Acid. Evaporated crude cleavages forthe different peptide moieties (23.7-79.0 mg, 0.03-0.09 mmol)were dissolved in individual vials using 0.5 mL of DMF, treatedwith DIEA (1 equiv, 5-15 μL, 0.03-0.09mmol), and stirred for5 min at rt. Second, D-lysergic acid (1.2 equiv, 11-29 mg,0.04-0.11 mmol) and HOAt (1.5 equiv, 6-18 mg, 0.05-0.14mmol) were dissolved in 1 mL of DMF (each reaction) andadded to every vial. Finally, DIPCDI was added (2 equiv, 9-27 μL, 0.06-0.18 mmol) and the set of reactions was stirred in aparallel synthesizer for 16 h. Multiple sample evaporationrendered the crude mixtures further purified by semipreparativeRP-HPLC.

Library Characterization. The purified compounds 1a-1sand 2a-2s were characterized by analytical RP-HPLC-MS,using a reverse-phase Symmetry C18 (5 μm, 3.9 mm � 150 mm)column and 1 mL/min flow. Elution system A: H2O-TFA,99.9:0.1; B: ACN-TFA, 99.9:0.1; gradient 0% B to 100% B in25 min. The purities of all library compounds (average g95%)were determined byUV absorption at 220 nm (HPLC-MS data,quantities, and purities of all library compounds are included inTable S1 in SI).

Synthesis of Biotin Ergopeptides (9-16). Rink-MBHA PS(1 g, loading: 0.56 mmol/g) was swollen with DCM (1 �1 min, 2 � 10 min) and DMF (5 � 1 min, 1 � 15 min) before

Chart 2. Structures of Biotin Ergopeptides 9-16

Article Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 1087

use. After washing, Alloc-L-Lys(Fmoc)-OH (3 equiv) wascoupled to the resin, using DIPCI (3 equiv) and HOBt (3 equiv)as a coupling system in DMF. The resin was washed and theFmoc group removed, yieldingAlloc-L-Lys(NH2)-AM-MBHA.At this point, the resin was split into two equal aliquots. Ac-Lys(Fmoc)-OH (3 equiv) and the Fmoc-Ahx-OH (3 equiv) wererespectively coupled on each one as previously described torender the resins 7 and 8. After that, Alloc group was eliminatedin both resins and biotin (3 equiv) was introduced using TBTU(3 equiv),DIEA (6 equiv), andHOBt (3 equiv) inDMF for 1 h atrt. After the Fmoc group removal, the resins were again dividedin two parts, obtaining at that point four different resins. Fromthis point onward, the attachment of the three different aminoacids (corresponding to the tripeptides of 1 and 2) was per-formed according to the procedure described previously. Thebiotin peptide moieties were cleaved following the B cleavageconditions, and the coupling of D-lysergic acidwas carried out asdescribed above. Evaporated crude cleavages for the differentbiotin peptide moieties (95-127 mg, 0.10-0.13 mmol) weredissolved in individual vials using 0.5 mL of DMF, treated withDIEA (1 equiv), and stirred for 5 min at rt. Second, D-lysergicacid (1.2 equiv) and HOAt (1.5 equiv) were dissolved in 1 mL ofDMF (each reaction) and added to every vial. Finally, DIPCDIwas added (2 equiv) and the set of reactions were stirred for 16 hat rt. Evaporation rendered the crude mixtures further purifiedby semipreparative RP-HPLC-MS under basic conditions(A: 20 mM NH4COOCH3, pH 9; B: ACN) using a reverse-phase X-Bridge C18 column (5 μm, 19 mm � 100 mm2) to yieldthe biotin ergopeptides 9-16.

Biotin Ergopeptides Characterization (9-16). Biotin ergopep-tides 9-16 were characterized by analytical RP-HPLC-MS andRP-HPLC using a reverse-phase XBridge C18 column (3.5 μm,4.6 mm � 50 mm2) at 2 mL/min (two separate elution solventsystems: (a) A, H2O-HCOOH (99.9:0.1); B, ACN-HCOOH(99.93:0.07). (b) A, H2O-TFA (99.9:0.1); B, ACN-TFA(99.9:0.1). The purities of 9-16 were determined by RP-HPLCat 220 nm UV absorption using as elution system b and agradient 5% B to 50%B in 4.5 min. All purities were confirmedto beg95%.HRMS spectra for 9-16were recorded confirmingthe identity of each biotin ergopeptides (Table 3).

Identity of compound 13 was confirmed by 1H NMR(400 MHz, DMSO-d6): 10.7 (s, 1H), 8.44 (s, 1H), 8.37 (s, 1H),8.29 (s, 2H), 8.17 (d, 2H), 7.90 (s,1H), 7.54 (d, 2H), 7.3-6.9 (m,7H), 7,30 (s,1H), 7.09 (s, 1H), 7.02 (s, 1H), 6.42 and 6.36 (s, 2H),6.26 (s, 1H), 4.98 (m, 1H), 4.64 (m, 1H), 4.4-3.8 (m, 7H), 3.00(dd, 1H), 2.78 (m, 1H), 2.59 (m, 2H), 2.51 (s, 3H), 2.3-1.24(m, 41H).

Biological Assays. Radioligand Binding Experiments. Gener-

al Procedure.Membrane suspensions from lamb brain striatumor D3R transiently transfected CHO cells were obtained byfollowing themethod previously described.35 Radioligand bind-ing assays using membrane suspensions (0.5 mg prot/mL deter-mined with bicinchoninic acid kits) were carried out at 22 �C in50 mM Tris-HCl buffer, pH 7.4 (see conditions below used foreach receptor). After radioligand incubation, free and mem-brane-bound ligand were separated by rapid filtration of 500 μLaliquots in a cell harvester through Whatman GF/C filtersembedded in 0.3% polyethylenimine (PEI) that were subse-quently washed for 5 s with 5 mL of ice-cold Tris-HCl buffer.The filters were incubatedwith 10mLof EcoscintH scintillationcocktail overnight at room temperature, and radioactivitycounts were determined using a scintillation counter with anefficiency of 62% for tritium labeled compounds and 99% forthe 125I-labeled compound.

Screening of the Library. Binding experiments of the wholelibrarywere performed at a concentration of 25 μMfor compounds1, 2, 1a-1s, 2a-2s and 9-16.

Dopamine D1, D2, and D3, Histamine H3, and MetabotropicGlutamate 1/5 Receptors. Membranes were incubated with0.9 nM [3H]-SCH23390 (85 Ci/mmol), 0.7 nM [3H]-YM09151-2 (85.5 Ci/mmol), 4 nM [3H]-raclopride (82.8 Ci/mmol), 2.5 nM[3H]-RAMH (34 Ci/mmol), or 27 nM [3H]-quisqualic acid(30.9 Ci/mmol) in 50 mM Tris-HCl buffer (pH 7.4) containing10 mMMgCl2 for 2 h in the absence or in the presence of testedcompounds. Nonspecific binding was measured in the presenceof 10 μM SCH23390, raclopride, RAMH, or quisqualic acid,respectively.

Somatostatin Receptors. Membranes were incubated with0.1 nM [125I]-Tyr11-somatostatin 14 (2,200 Ci/mmol) in 50 mMTris-HCl buffer (pH 7.4) containing 10mMMgCl2 for 3 h in theabsence or in the presence of tested compound. Nonspecificbinding was measured in the presence of 100 nM somatostatin.

Adenosine A1 and A2A Receptors.Membranes were incubatedwith 1.0 nM [3H]-R-PIA (30.5 Ci/mmol) or 1.6 nM [3H]-ZM241385 (27.4 Ci/mmol) in 50 mM Tris-HCl buffer (pH 7.4)containing 10 mM MgCl2 and 0.2 U/mL ADA for 2 h in theabsence or in the presence of tested compounds. Nonspecificbinding was measured in the presence of 10 μMDPCPX or ZM241385, respectively.

Cannabinoid CB1 Receptor. Membranes were incubated insiliconated tubes with 0.5 nM [3H]-CP 55940 (144 Ci/mmol) in

Figure 3. Competition curves of dopamine receptors antagonists binding versus increasing concentrations of compound 13. Competitionexperiments of 0.9 nMD1Rantagonist [3H]-SCH23390 (a) or 0.7 nMD2Rantagonist [3H]-YM09151-2 (b) versus increasing concentrations ofcompound 13were performed with brain striatal membranes (0.5 mg prot/mL) as indicated in the Experimental Section. Data are representedas means ( SD from a representative experiment (n = 3) performed in triplicate.

Table 2. KD Values of Selected Biotin Ergopeptides at D1R and D2R

D1R D2R

KDB1 (nM) KDB2 (μM) KDB1 (nM) KDB2 (μM)

9 107( 9 3.2( 0.3 260( 60 7( 2

12 2100( 300 >50 2000( 400 23( 8

13 56( 6 1.5 ( 0.2 17( 5 9( 3

14 490( 80 >50 160( 50 10( 4a KDB1 and KDB2 are respectively the equilibrium dissociation con-

stants of the high and low binding affinities of ergopeptides to thedimeric D1R and D2R.

1088 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 Vendrell et al.

50mMTris-HCl buffer (pH7.4) containing 10mMMgCl2 and 1mg/mL fatty acid-free bovine serum albumin (BSA) for 2 h inthe absence or in the presence of tested compounds. In this case,filters were presoakedwith buffer containingBSA, andBSAwasmaintained in the washing medium. Nonspecific binding wasmeasured in the presence of 10 μM CP55940.

KD Determination. Competition experiments of 0.9 nM [3H]-SCH23390, 0.7 nM [3H]-YM09151-2, or 4 nM [3H]-raclopridebinding versus increasing concentrations of compounds 9, 12,

13, or 14 were performed by incubating membranes under thesame conditions as described above for D1R, D2R, or D3Rbinding. Nonspecific binding was determined as previouslyoutlined. Radioligand displacement curves were analyzed bynonlinear regression using the commercial program GRAFIT(Erithacus Software, Surrey, UK) by fitting the specific bindingdata to the mechanistic two-state dimer receptor model.35,36 Tocalculate the macroscopic equilibrium dissociation constants,the equations used for a competition binding experiment werededuced by Casad�o.35 Goodness-of-fit was tested following thereduced χ2 value given by the nonlinear regression programGRAFIT. A modified F test was used to analyze whether the fitto cooperativity model significantly improved upon the fit tononcooperative model, and p<0.05 was taken as a criterion ofsignificance; when no significant improvement over the non-cooperative model was detected, the p values were >0.30.

CellCulture,TransientTransfection, andProteinDetermination.

Chinese hamster ovary (CHO) cells were cultured in MEMR medium without nucleosides supplemented with 100 U/mLpenicillin/streptomycin and 10% (v/v) heat inactivated fetalbovine serum (FBS). CHO cells were maintained at 37 �C in ahumidified atmosphere of 5% CO2 and were passaged whenthey were 80-90% confluent, i.e., approximately twice a week.CHO cells were transiently transfected with 2 μg cDNA corre-sponding to human D1R, D2R, or D3R by the ramified PEImethod. Cells were incubated for 4 h with the correspondingcDNA together with ramified PEI (5 mL/mg cDNA of 10 mMPEI) and 150 mM NaCl in a serum-starved medium. After 4 h,the medium was changed to a fresh complete culture medium.Forty-eight h after transfection, cells were washed twice in quicksuccession in HBSS (Hanks’ balanced salt solution: 137 mM

Figure 4. Functional characterization of the biotin ergopeptide 13. CHO cells expressing D1R (a and c) or D2R (b and d) were culturedin serum-free medium for 16 h prior to the addition of any ligand. Cells were treated (or not) with 50 μM of the D1R antagonist SCH23390(a and c) or the D2R antagonist raclopride (b and d). After 5 min, increasing concentrations of compound 13, D1R agonist SKF 81297, or theD2R agonist quinpirole were added for further 5 min of incubation, and ERK1/2 phosphorylation (a and b) or P-Ser473Akt (c and d) weredetermined as indicated in the Experimental Section. Results are expressed as means( SEM of four independent experiments. Student’s t-testfor unpaired samples showed significant increases (a and b) or decreases (c and d) over basal (not treated cells, *p < 0.05, **p < 0.01) orsignificant decreases (a and b) or increases (c and d) respect to cells stimulated with the same concentration of compound 13 in the absence ofantagonist (#p < 0.05, ##p < 0.01).

Figure 5. Competition curve of a D3R antagonist binding versusincreasing concentrations of compound 13. Competition experi-ments of 4 nM D3R antagonist [3H]-raclopride versus increasingconcentrations of compound 13 were performed with membranes(0.5 mg prot/mL) from CHO cells transiently transfected with 2 μgof the cDNA corresponding to D3R, as indicated in the Experi-mental Section. Data are represented as means ( SD from arepresentative experiment (n = 3) performed in triplicate.

Article Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 1089

NaCl, 5 mM KCl, 0.34 mM Na2HPO4 3 12H2O, 0.44 mMKH2PO4, 1.26 mM CaCl2 3 2H2O, 0.4 mM MgSO4 3 7H2O,0.5 mM MgCl2, 10 mM HEPES, pH 7.4) supplemented with0.1% glucose (w/v), detached by gently pipetting and resus-pended in the same buffer. To control the cell number, sampleprotein concentration was determined using a Bradford assaykit using BSA dilutions as standards.

ERK and Akt Phosphorylation Assay. Transfected CHO cellswere cultured in serum-free medium for 16 h before the additionof any agent. Cells were treated or not with 5 or 50 μM of theD1R antagonist SCH 23390 or the D2R antagonist raclopride.After 5 min, increasing concentrations of compound 13, 1 μMD1R agonist SKF 81297, or 1 μMD2R agonist quinpirole wereadded for further 5 min incubation. Cells were rinsed with ice-cold phosphate-buffered saline and lysed by the addition of500 μL of ice-cold lysis buffer (50 mMTris-HCl pH 7.4, 50 mMNaF, 150 mM NaCl, 45 mM β-glycerophosphate, 1% TritonX-100, 20 μM phenyl-arsine oxide, 0.4 mM NaVO4, and pro-tease inhibitor cocktail). The cellular debris was removed bycentrifugation at 13000g for 5 min at 4 �C, and the protein wasquantified by the bicinchoninic acidmethod using bovine serumalbumin dilutions as standard. To determine the level of ERK1/2 or Akt-phosphorylation, equivalent amounts of protein (10 μg)were separated by electrophoresis on a denaturing 7.5% SDS-polyacrylamide gel and transferred onto PVDF-FLmembranes.Odyssey blocking buffer was then added, and themembranewasrocked for 90 min. The membranes were then probed with amixture of antiphospho-ERK1/2 antibody (1:2500) or anti-P-Ser473Akt antibody (1:2500) and anti-ERK1/2 antibody thatrecognizes both phosphorylated and nonphosphorylated ERK1/2(1:40000) for 2-3 h. Bands were visualized by the addition of amixture of IRDye 800 (antimouse) antibody (1:10000) andIRDye 680 (antirabbit) antibody (1:10000) for 1 h and scannedby the Odyssey infrared scanner. Bands densities were quanti-fied using the scanner software, exported to Excel. The level ofphosphorylated ERK1/2 isoforms or P-Ser473Akt was normal-ized for differences in loading using the total ERK protein bandintensities.

Acknowledgment. We acknowledge the technical help ob-tained from Jasmina Jim�enez (Molecular Neurobiology La-boratory, University of Barcelona), from Serveis Cientı́fic-T�ecnics of University of Barcelona for their support in theHRMS, and Marı́a Macı́as (Institute for Research inBiomedicine) for their support in NMR analysis. This workwas partially supported by grants from SpanishMinisterio deCiencia e Innovaci�on MICINN (SAF2008-00146, SAF2008-03229-E, SAF2009-07276, BQU2006-03794, CTQ2005-00315/BQU,CTQ2008-00177, SAF2005-00170, and SAF2006-05481),grant 060110 fromFundaci�oLaMarat�o deTV3,Generalitat deCatalunya, CIBER-BBN, Networking Centre on Bioengi-neering, Biomaterials, and Nanomedicine Institute for Re-search in Biomedicine, and the Barcelona Science Park. P.J.M. is a Ramon y Cajal Fellow.

Supporting Information Available: Additional syntheticschemes, tables with chemical structures, and complete

characterization data for the ergopeptide library, radioligandbinding assays of 1c, 1f, 2c, and 2f (A1R and A2AR) and 9-16

(D1R, D2R, A1R, and A2AR), competition curves of 9, 12, and14 at D1R and D2R and binding experiments of 13 at differentGPCRs. This material is available free of charge via the Internetat http://pubs.acs.org.

References

(1) Vendrell, M.; Angulo, E.; Casad�o, V.; Lluis, C.; Franco, R.;Albericio, F.; Royo, M. Novel ergopeptides as dual ligands foradenosine and dopamine receptors. J. Med. Chem. 2007, 50, 3062–3069.

(2) Maggio, R.; Aloisi, G.; Silvano, E.; Rossi, M.; Millan, M. J.Heterodimerization of dopamine receptors: new insights intofunctional and therapeutic significance. Parkinsonism Relat. Dis-ord. 2009, 15 (S4), S2–S7.

(3) So, C.H.; Verma,V.; O’Dowd,B. F.;George, S.R.Desensitizationof the dopamine D1 and D2 receptor hetero-oligomer mediatedcalcium signal by agonist occupancy of either receptor. Mol.Pharmacol. 2007, 72, 450–462.

(4) So, C. H.; Varghese, G.; Curley, K. J.; Kong, M. M.; Alijaniaram,M.; Ji, X.; Nguyen, T.; O’Dowd, B. F.; George, S. R. D1 and D2

dopamine receptors form heterooligomers and cointernalize afterselective activation of either receptor. Mol. Pharmacol. 2005, 68,568–578.

(5) Bhushan, R. G.; Sharma, S. K.; Xie, Z.; Daniels, D. J.; Portoghese,P. S. A bivalent ligand (KDN-21) reveals spinal delta and kappaopioid receptors are organized as heterodimers that give rise todelta(1) and kappa(2) phenotypes. Selective targeting of del-ta-kappa heterodimers. J. Med. Chem. 2004, 47, 2969–2972.

(6) Waldhoer, M.; Fong, J.; Jones, R. M.; Lunzer, M. M.; Sharma,S. K.; Kostenis, E.; Portoghese, P. S.; Whistler, J. L. A hetero-dimer-selective agonist shows in vivo relevance of G protein-coupled receptor dimers. Proc. Natl. Acad. Sci. U.S.A. 2005, 102,9050–9055.

(7) Adam, G. C.; Sorensen, E. J.; Cravatt, B. F. Proteomic profiling ofmechanistically distinct enzyme classes using a common chemotype.Nature Biotechnol. 2002, 20, 805–809.

(8) Speers, A. E.; Adam, G. C.; Cravatt, B. F. Activity-based proteinprofiling in vivo using a copper(I)-catalyzed azide-alkyne [3 þ 2]cycloaddition. J. Am. Chem. Soc. 2003, 125, 4686–4687.

(9) Walsh, D. P.; Chang, Y. T. Chemical genetics. Chem. Rev. 2006,106, 2476–2530.

(10) Middleton, R. J.; Briddon, S. J.; Cordeaux, Y.; Yates, A. S.; Dale,C. L.; George, M. W.; Baker, J. G.; Hill, S. J.; Kellam, B. Newfluorescent adenosine A1-receptor agonists that allow quantifica-tion of ligand-receptor interactions in microdomains of singleliving cells. J. Med. Chem. 2007, 50, 782–793.

(11) Lokey, R. S. Forward chemical genetics: progress and obstacles onthe path to a new pharmacopoeia. Curr. Opin. Chem. Biol. 2003, 7,91–96.

(12) Middleton, R. J.; Kellam, B. Fluorophore-tagged GPCR ligands.Curr. Opin. Chem. Biol. 2005, 9, 517–525.

(13) Gellerman, G.; Elgavi, A.; Salitra, A.; Kramer, M. Novel Glybuilding units for backbone cyclization: synthesis and incorpora-tion into model peptides. J. Pept. Res. 2001, 57, 277–291.

(14) Vendrell, M.; Ventura, R.; Ewenson, A.; Royo, M.; Albericio, F.N-[Chloro(dimethylamino)methylene]-N-methylmethanaminiumchloride (TMUCl Cl), the reagent of choice for the solid-phasesynthesis of anilides. Tetrahedron Lett. 2005, 46, 5383–5386.

(15) Baraldi, P. G.; Tabrizi, M. A.; Gessi, S.; Borea, P. A. Adenosinereceptor antagonists: translatingmedicinal chemistry and pharma-cology into clinical utility. Chem. Rev. 2008, 108, 238–263.

(16) Agnati, L. F.; Ferr�e, S.; Lluis, C.; Franco, R.; Fuxe, K. Molecularmechanisms and therapeutical implications of intramembranereceptor/receptor interactions among heptahelical receptors with

Table 3

compd tR (min) molecular formula Mcalc Mexp (HRMS) purity (220 nm) (%) mg

9 3.33 C57H78N12O10S 1122.5685 1123.5747 95 12.6

10 3.68 C57H78N12O10S 1122.5685 1123.5747 99 14.3

11 3.20 C59H81N13O11S 1179.5899 1180.5961 99 14.5

12 3.58 C59H81N13O11S 1179.5899 1180.5963 98 13.6

13 3.88 C62H79FN12O10S 1202.5747 1203.5815 99 14.7

14 3.85 C62H79FN12O10S 1202.5747 1203.5804 94 15.5

15 3.81 C64H82FN13O11S 1259.5961 1260.6015 95 12.9

16 3.77 C64H82FN13O11S 1259.5961 1260.6016 96 25.1

1090 Journal of Medicinal Chemistry, 2011, Vol. 54, No. 4 Vendrell et al.

examples from the striatopallidalGABAneurons.Pharmacol. Rev.2003, 55, 509–550.

(17) Franco, R.; Casad�o, V.; Mallol, J.; Ferr�e, S.; Fuxe, K.; Cort�es, A.;Ciruela, F.; Lluis, C.; Canela, E. I. Dimer-based model for heptas-panningmembrane receptors.Trends Biochem. Sci. 2005, 300, 360–366.

(18) Bayer, E. A.; Wilchek, M. Application of avidin-biotin techno-logy to affinity-based separations. J. Chromatogr. 1990, 510, 3–11.

(19) Elia,G. Biotinylation reagents for the study of cell surface proteins.Proteomics 2008, 8, 4012–4024.

(20) Bratthauer, G. L. The avidin-biotin complex (ABC) and otheravidin-biotin bindingmethods.MethodsMol. Biol. 1999, 115, 203–214.

(21) Smith, D. J.; Shappell, N. W. Technical note: epimerization ofergopeptine alkaloids in organic and aqueous solvents. J. Anim.Sci. 2002, 80, 1616–1622.

(22) Chausmer, A. L.;Katz, J. L. Comparison of interactions ofD1-likeagonists, SKF 81297, SKF 82958, and A-77636, with cocaine:locomotor activity and drug discrimination studies in rodents.Psychopharmacology 2002, 159, 145–153.

(23) Levant, B.; Grigoriadis, D. E.; De Souza, E. B. Characterization of[3H]-quinpirole binding toD2-like dopamine receptors in rat brain.J. Pharmacol. Exp. Ther. 1992, 262, 929–935.

(24) Beaulieu, J.; Gainetdinov, R. R.; Caron, M. G. The Akt/GSK-3signaling cascade in the actions of dopamine. Trends Pharmacol.Sci. 2007, 28, 166–172.

(25) Beaulieu, J.; Gainetdinov, R. R.; Caron, M. G. Akt/GSK3 signal-ing in the action of psychotropic drugs. Annu. Rev. Pharmacol.Toxicol. 2009, 49, 327–347.

(26) Marcellino, D.; Ferr�e, S.; Casad�o, V.; Cort�es, A.; Le Foll, B.;Mazzola, C.; Drago, F.; Saur, O.; Stark, H.; Soriano, A.; Barnes,C.;Goldberg, S. R.; Lluis, C.; Fuxe,K; Franco,R. Identification ofdopamine D1-D3 receptor heteromers. Indications for a role ofsynergistic D1-D3 receptor interactions in the striatum. J. Biol.Chem. 2008, 283, 26016–26025.

(27) Ferrada, C.; Moreno, E.; Casad�o, V.; Bongers, G.; Cort�es, A.;Mallol, J.; Canela, E. I.; Leurs, R.; Ferr�e, S.; Lluı́s, C.; Franco, R.Marked changes in signal transduction upon heteromerization of

dopamine D1 and histamine H3 receptors. Br. J. Pharmacol. 2009,157, 64–75.

(28) Ferrada, C.; Ferr�e, S.; Casad�o, V.; Cort�es, A.; Justinova, Z.;Barnes, C.; Canela, E. I.; Goldberg, S. R.; Leurs, R.; Lluis, C.;Franco, R. Interactions between histamine H3 and dopamine D2

receptors and the implications for striatal function. Neuropharma-cology 2008, 55, 190–197.

(29) Cabello, N.; Gandı́a, J.; Bertarelli, D. C.; Watanabe, M.; Lluı́s, C.;Franco, R.; Ferr�e, S.; Luj�an, R.; Ciruela, F. Metabotropic gluta-mate type 5, dopamine D2 and adenosine A2A receptors formhigher-order oligomers in living cells. J. Neurochem. 2009, 109,1497–1507.

(30) Kearn, C. S.; Blake-Palmer, K.; Daniel, E.; Mackie, K.; Glass, M.Concurrent stimulation of cannabinoid CB1 and dopamine D2

receptors enhances heterodimer formation: a mechanism for receptorcross-talk?Mol. Pharmacol. 2005, 67, 1697–1704.

(31) Rocheville,M.; Lange, D. C.; Kumar, U.; Patel, S. C.; Patel, R. C.;Patel, Y. C. Receptors for dopamine and somatostatin: formationof hetero-oligomers with enhanced functional activity. Science2000, 288, 154–157.

(32) Vazquez, J.; Qushair, G.; Albericio, F. Qualitative colorimetrictests for solid phase synthesis.Methods Enzymol. 2003, 369, 21–35.

(33) Strohmeier, G.A.; Haas,W.;Kappe, C. O. Synthesis of functionalized1,3-thiazine libraries combining solid-phase synthesis and post-clea-vage modification methods. Chem.;Eur. J. 2004, 10, 2919–2926.

(34) Attardi, M. E.; Porcu, G.; Taddei, M. Malachite green, a valuablereagent to monitor the presence of free COOH on the solid-phase.Tetrahedron. Lett. 2000, 41, 7391–7394.

(35) Casad�o, V.; Ferrada, C.; Bonaventura, J.; Gracia, E.; Mallol, J.;Canela, E. I.; Lluı́s, C.; Cort�es, A.; Franco, R. Useful pharmaco-logical parameters for G-protein-coupled receptor homodimersobtained from competition experiments. Agonist-antagonistbinding modulation. Biochem. Pharmacol. 2009, 78, 1456–1463.

(36) Casad�o, V.; Cort�es, A.; Ciruela, F.; Mallol, J.; Ferr�e, S.; Lluı́s, C.;Canela, E. I.; Franco, R. Old and newways to calculate the affinityof agonists and antagonists interacting with G-protein-coupledmonomeric and dimeric receptors: the receptor-dimer cooperativityindex. Pharmacol. Ther. 2007, 116, 343–354.